Existing research supports a role for the intrauterine environment in modifying risk for obesity in offspring. Maternal obesity and relatively high glucose concentrations are two commonly cited predictors of obesity in the next generation, potentially via the delivery of excess fuel across the placenta to the developing fetus, which in turn, may program fetal metabolism in a manner that promotes excess weight gain in the future. Although a number of lifestyle interventions to improve maternal and infant outcomes have previously been attempted, results are often modest at best. Furthermore, few studies have proved to be efficacious among low income African American women who have high risk for the transmission of obesity to future generations. Research previously conducted by the PI (K01DK090126; DRC pilot award), has discovered that late-night consumption of high glycemic index (GI) foods is very prevalent in late pregnancy among African American women who are obese. Consistent with the existing literature describing adverse metabolic consequences of shift-work and late-night eating, this diurnal pattern of intake was associated with reduced glucose tolerance in the mothers, and a number of infant outcomes that are themselves associated with future obesity. The overall goal of this study is to follow-up this previous research by examining the feasibility of developing an intervention to reduce late-night intake of high-GI foods and beverages among obese African American women, in order to improve maternal glucose tolerance and reduce excess fetal growth in late gestation. This goal will be achieved with a two-phase study. First, focus groups will be conducted with pregnant African American women who are obese, in order to identify reasons for late-night consumption of high-GI foods, barriers against changing this behavior, and perceptions of an intervention to reduce late-night high-GI intake. Second, a randomized control trial (RCT) design will be used to examine the feasibility of conducting an intervention to reduce late-night high-GI consumption among obese women during the third trimester. The intervention will provide lower-GI alternatives to replace the high-GI foods and beverages that are typically consumed by the target population. Changes in free-living glycemic variability, glucose tolerance, and fetal abdominal growth, during the intervention will be measured and compared to those of women in a control (i.e. standard care) group. At the conclusion of this study, it is anticipated that suffiient preliminary data will have been obtained with which to inform a larger and more comprehensive RCT. Furthermore, this study will support the PI's transition to independence by providing an opportunity to experimentally examine the biobehavioral mechanisms underlying the development of obesity following intrauterine exposure to maternal obesity and/or high glucose concentrations.